The average resonance parameters recommended for RIPL-2 were prepared on the basis of the evaluations performed by the Obninsk group, taking into account the analysis of discrepancies between similar evaluations of other groups. Good agreement was found for Г7 among the three RIPL-1 files (Obninsk, Mughabghab and Beijing) [5]. However, the comparison is less fa­vorable for the neutron strength functions, especially for those cases with a large number of resonances. The revised average resonance parameters were obtained for 20 additional nuclei for which the data on resolved resonance parameters are available in the Sukhoruchkin compilation [9], bringing the total number of D05S in RIPL-2 to 301. Generally, the accuracy of these additional data is rather poor due to the low number of resonances available for analysis. New evaluations of the average parameters for p-wave neutron resonances, prepared by the Obninsk group, have been included in the up­dated version of the RIPL-2 file. These resonances provide a good check of consistency since they are known to be about a factor of 3 smaller than the s-wave spacings, which is particularly relevant for magic nuclei.

Careful attention was paid to the estimation of uncertainties for the recommended parameters, based on experienced guesswork of systematic errors beside statistical uncertainties.